Hmmm… The description certainly fits. Just by eye-balling the graphs, they look very different from what I got, but I guess that’s just the expected result of running rbinom about a 6 million times. With a smaller simulation, it might not have been so apparent. Also, that’s what you get for skipping the maths and vibing the code without thinking too much about the details. Well, at least i got this far with absolutely minimal effort. :D

It appears that I need to switch to a better distribution. Thanks for looking into this mystery!